UNIT 1 ANALYTIC FUNCTIONS

Structure Page No.

! 1.I Introduction
! Objectives
1.2 Basic Concepts
1.3 Functions of a Complex Variable
Mappings
Limits
Continuity
Differentiation
1.4 Cauchy-Riemann Equations
1.5 Analyticity
1.6 Harmonic Functions
1.7 Summary
1.8 Solutions/Answers

1 . INTRODUCTION

There are equations such as x 2 + 3 = 0, x - 1Ox + 40 = 0 , which do not have a root in

.
the real number system R There does not exist any real number whose square is a
negative real number. If the roots of such equations are to be determined then we need
to introduce another number system called complex number system. Complex
numbers can be defined as ordered pairs (x, y) of real numbers and represented as
points in the complex plane, with rectangular coordinates x and y .

In Sec. 1.2, we shall review the algebraic and geometric structure of the complex
number system. Sec. 1.3 introduces you to the functions of a complex variable.
Further, we shall study the concepts of limits, continuity and differentiability of the
functions of a complex variable in this section. In Sec. 1.4, you will study about
Cauchy-Riemann equations which under certain conditions provide the necessary and
sufficient condition for the differentiability of a function of a complex variable at a
point. A very important concept of analytic functions which is useful in many
application of the complex variable theory is discussed in Sec. 1.5. Lastly, in Sec. 1.6,
we shall introduce harmonic functions which play an important role in applied
mathematics. The temperature T(x, y) in thin plates in the xy -plane and a function
V(x, y) denoting an electrostatic polential in a region free of charges are some of the
examples of harmonic function.

Objectives
After studying this unit you should be able to
define a complex number and represent it as a point in the complex plane;
define a function of complex variable and also the limit, continuity and
differentiability of a function;
use the Cauchy-Riemann equations to prove the differentiability of a function of a
complex variable at a point;
define the analyticity of a function at a point and discuss it.

1.2 BASIC CONCEPTS

This section introduces you to the concept of complex numbers. It gives you an idea of
Complex Analysis how an ordered pair (x, y) of real numbers can be interpreted as a point in the i
complex plane with rectangular coordinates x and y . 1:
Read Secs. (1)-(10) of Chapter 1 of the book from pages 1-32. Carefully go
The book mentioned through all the solved examples in each of the ten sections of the chapter and
everywhere refers to try to do the exercises given at the end of each section.
'Complex Variables and
Applications' (seventh
edition) by J. W. Brown This chapter will help you to understand the definition of complex numbers, their
and R.V. Churchill. algebraic properties of addition and multiplication, geometrical representation of a
complex number as a point in the plane, polar and exponential form of a complex
numbers and their roots and powers. Carefully go through this chapter as all the
results and definitions given there would be used frequently in the chapters to follow.
You have to make yourself comfortable with all the notions of Sec. 10. In case you are
already familiar with these concepts, a quick glance through this chapter will help you
recall these concepts. Chapter 1 of the book is introductory. For the rest of the

tI sections in this unit, we shall confine ourselves to Chapter 2 of the book.

Before proceeding further, we would like to make the following remarks.

Remark 1: Every open and convex set (any pair of two points of it can be joined by a
line segment lying in the set) is a connected set but every open connected set may not
be convex (see Fig. 16 page 3 1 of the book).

Remark 2: Union of two open connected sets may not be an open connected set. For
example, see Exercise 5 on page 32 of the book, S = (z :I z 1 < 1 or I z - 2 1 < I). Here
z, cannot be joined to z, by a polygonal line without crossing the boundary (see
Fig. 1
Fig. 1). Thus S is not connected but is open.

1.3 FUNCTIONS OF A COMPLEX VARIABLE

In this section we shall introduce you to functions of a complex variable and develop
the concepts of its mappings, limits, continuity and differentiability.

You can start with the following:

Read Sec. 11, Chapter 2 of the book from pages 33-35. Go through all the
four examples and do the exercises l(c), l(d), 4 on page 35 of the book.

1.3.1 Mappings
To display information about the function w = f (z) ,by indicating pairs of
corresponding points z = (x, y) and w = (u, v) , it is simpler to draw the z and w
planes separately and this is referred to as a mapping or transformation. Through this
we can understand the concept of mapping and get an idea of finding the image in
w-plane of any region in z-plane under the transformation w = f (2).

Read Secs. 12 and 13, Chapter 2 of the book from pages 36-42. Do the
exercises 4,5 on pages 42,43 of the book.

We now move on to the concept of limit of a function of a complex variable.

1.3.2 Limits Analytic Functions

Read Secs. 14 -15, Chapter 2 of the book from pages 43-48.

Sec. 14 introduces you to the finite limit of a function. Sec. 15 gives you some
theorems on limits. These theorems help you in establishing a connection between
limits of functions of a complex variable and limits of real-valued functions of two
real variables. Here in addition to the results given in Sec. 15 we would like to add
another result on the limit of a function of a complex variable which we are giving in
the form of a theorem as follows:

Theorem 1: If f(z) has a finite limit at z, then f(z) is a bounded function in some
deleted neighbourhood of z, .
Since lim f (z) exists as a finite complex number, we can take it as some complex
z+zo

I I
number w, and write f(z) - w,l < E whenever, 0 < z - z,l < 6 . Hence, for any z
in the neighbourhood.of z, , we find that

I
where M = E + wOI is some finite number. Therefore f (z) is bounded in some
neighbourhood of z, .
***
Let us once again consider Theorem 2 on page 47 of the book. Property (9) has been
proved using Theorem 1 on page 46. Let us now see how this result can be proved
directly by using the definition of the limit of a function of a complex variable.
To prove the result using the definition of limit we write, for given E > 0 3 6,, 6, > 0
such that
( f ( z ) - w , ( < E , whenever o<Iz-z, 1<6,,
and (F(z)-W, I < & , whenever o < ( z - z , ( < 6,.

Now, I f (z) F(z) - w ,W0( =( f (z) (F(z) - W, ) + W, (f (z) - w ,) I

s f(z)(F(z)-W,)
I ~(z)IIF(z)-w, II
+ W,(f(z)-w,)(
+. woI (f(z)-w,I.

Since lim f(z) exists and is finite, f(z)

z-+ZO
I ISM, for a finite constant M by Theorem 1
above. Choosing 6 = min (6, ,ti,) , we get
I f ( z ) ~ ( z ) - w , ~ , II M E + I W , ) E = E , , for O < ) Z - Z , I< t i .
I
Thus, f (z) F(z) - w, W, < E, I
This establishes Property (9) of Theorem 2 on page 47 of the book.

Now to get introduced to the limit of a function at a point at infinity and infinite limits
you may read the following:

Read Sec. 16, Chapter 2 of the book from Pages 48-51.

Before we ask you to solve some problems we give you some more examples to have
a better grasp of the concepts covered in Secs. 14-16.
Complex Analysis Example 1: Show that lim(2x + iy2) = 4i
z+2i
Solution: For each positive number E we must find a positive number
6 such that
12x+iy2 -4il< E whenever, 0 < ( z - 2 i l < 6 . (1)
To do this we write

Thus, the first of inequality (1) will be satisfied if

E E
21x1<- and l y - 2 1 ) y + 2 ) < - .
2 2
1
The first of these inequalities is, of course, satisfied if x ( < E / 4 . To establish
conditions on y such that the second one holds, we restrict.y so that
Re = Real part
Im = Imaginary part
Iy-21 <min -, 1
rl"0 1 . Thenclearly, Iy-21 < l a n d w e findthat

E E
Therefore, ( y - 2 / I y + 2 1 < -.5= - . Hence, fiom the two conditions;
10 2
E
the appropriate value of 6 is 6 = min

1
Example 2: Show that lim 7= 0
z+m z
Solution: For a given real number E > 0 ,we want to determine a real
number 6 > 0 such that
X
1 --1 )
< E whenever z > -
1
6
1 1
I1 1
Now, T 1 < ~ 3 ) z I > - . ~ h u s w e c a n t a k e 6 6 & . Withthischoi~g.of6,we
&
find
I -$1 < E whenever I z 1 > -61.

Yf Example 3: Show that the following limits do not exist

z [ ~ e -z 1 m z I 2
a) lim - b) lim
z+o lzI
l
ITrn
z-ro
z12
Solution: We are going to use a simple fact that limit, if exists, is independent of
Y' choice of paths or the approaches the variable z takes to reach the point (at which you
intend to calculate the limit).
a) We choose a path such that x + 0 is followed by y + 0 (see Fig. 2).
We get (when x + 0 )
(b)

Fig. 2

andwhen y + 0 ' %-ti

' IYI
andwhen y + o - , %+-i. [Fig.2(b:)]
IY I
Thus limit does not exist.
b) Consider any straight line path y = mx . Taking limit along this path, we have Analytic Functions

[ ~ e -zh z 1 2 (x-Y)' =lim ( ~ - m )x2

~ -
-
(1 -m)2
lim = lim
z+o
lzlZ z+o x2+Y2 x+0(1+m2)x2 (l+m2)
the limit depends on m and is not unique so, the limit does not exist.
-X-

You may now try the following exercises:

Do the exercises 4,5 on page 53 and exercises 9,11,13 on page 54 of the

book.

With this background you may now move on to the next section.

1.3.3 Continuity
This section introduces you to the continuity of the function of a complex variable at a
point.

Read Sec. 17, Chapter 2 from pages 51-53 of the book.

After you have gone through Sec. 17 of the book we would like you to notice that
result given by Eqn.(6) of Theorem 2 on page 53 can be Stated as follows:
If f(z) is continuous in a closed region R , then it is bounded in R , that is,
If(z1 s M for all z in R and for some M > 0 .

Further, we would like to add the following remark.

Remark: In the definition of continuity given by Eqns.(l)-(3) on page 5 1 of the book,

if both f(z,) and lim f(z) = Z (say) exist, but f(z,) # Z ,then point z, is called a
z+zo
point of removable discontinuity. In this case, we can redefine the function f (z) at
z, such that f(z,) = Z and the function can be made continuous at z, . This can also
z2 - 4
be done even if f(z,) is not defined there. For instance, consider f(z) = -.
z+2
Here f (z) is not defined at z = -2 . If we redefine f (-2) as
(z - 2)(z + 2)
f (-2) = lim f (z) = lim = -4,
z+-2 z+-2 (z + 2)
then the function becomes continuous at z = -2. So that z = -2 is a removable
discontinuity. Thus, function defined as

[-4 , z=-2
is continuous for all z .

Let us consider the following examples.

Example 4: Show that the functions

a) f ( z ) = e x cosy+iex siny b) f (z) = sin x cosh y + i cos x sinh y
C) f(z)=xy2 +i(2x -y)
are continuous for all z .
Complex Analysls Solution: a) .
We can write z = x + iy Therefore
f(z)=eXcosy+iex s i n y = u + i v = w
Since the real valued functions u and v of two real variables x and
y are continuous at each (x, y) in R 2 , the function f(z) is
continuous for all z .
b) f(z)=sinxcoshy+icosxsinhy = u + i v .
f(z) is continuous for all z by the same argument as in a) above.
c) Likewise, the function
f(z) =xy2 + i(2x - y)
is continuous everywhere in the complex plane as the component
functions are polynomials in x and y are therefore continuous at each
point (x, y) E R .

Example 5: Show that the function

[
0 ,z=o
is not continuous at z = 0 .
Solution: If f(z) is continuous at z = 0 then
Im z
lim- = lim must exist.
1
z+o z ( (x, Y,-b(O. 0) ,/+yf

Let us consider the path y = mx then

mx m
1im
(x. Y)+(O. 0) Jm =lim
x-bo ,/=
= ,/=
*

But m is a variable i.e., limiting value depends on the path chosen. Hence lim f(z)
z+o
does not exist. Therefore the function is not continuous at z = 0 .
-X-

How about trying these exercises now.

El) Show that the function

a) f(z)=1zl2 b) f(z)=exY+isin(xZ -2xy3)
are continuous for all z .
Find the value of f (i) so that the function f (z) =
3z4 -2z3 + 8zz -2z+5 is
E2)
z-i
continuous at z = i .

E3) Find the points in the complex plane at which the following functions
are not continuous.

Let us now move on to the derivative of a function of a complex variable.

1.3.4 Differentiation

Read Secs. 18-19, Chapter 2 from pages 54-59 of the book.

 *

Secs. 18 and 19 of the book introduce you to the concept of derivative of a function Analytic Functions
f(z) at a point in the complex plane. After going through these sections you must
have realised:
i) A function which is not continuous at a point z = z, cannot be differentiable at
z=zo.
ii) A function which is continuous at a point z = z,, may or may not be
differentiable at z = z, .
iii) The rules of differentiation of a function of a real variable x hold also for a
function of a complex variable z .
***
The following examples may be of help to you for doing the exercises, so go through
them before trying problems.

Example 6: Show that the function f (z) = Z is continuous at the point z = 0 but not
differentiable at z = 0 . Is this function continuous and differentiable elsewhere?
Solution: Let w = f(z) = u(x, y) + iv(x, y) .
f(z) = H = x - iy gives u(x, y) = x, v(x, y) = -y .
Now lim u(x, y) = lim x=O
(x* Y)+(O* 0) (Xu Y)+(OP 0)

and lim lim (-y) = O

(x* Y)+(O* 0) v(x* y) = (x, y)+(O, 0)

therefore, lim f (z) = 0 + 0 = 0 and also f (0) = 0

z+o
.
Hence, lirn f (z) = f (0) 3 f is continuous at z = 0 .
z+o

In fact, f (z) = ;i is continuous everywhere. For if z = z, be any point then

If(z)-f(z,)I = IZ-8, I = Iz-Z, 1.
Thus for given E > 0 such that
(f(z)-f(z,)I < &,wehave Iz-z, I < 6 = E .
To check the differentiability of f(z)
let Aw = f (z + Az) - f (z) .
-
Aw f(0 + Az) - f(0) - _--- - Az
f(Az)
At z = o , -=
A2 Az Az Az
Aw
If the limit of - exists, it may be found by letting the point Az = (h, Ay)
Az
approach the origin in the Az -plane in any manner. In particular, when Az
approaches the origin horizontally through the point (Ax, 0) on the real axis (see
Fig. 3

Aw Az
Thus, -=-=I.
Az Az
When Az approaches origin vertically through Ay -axis, we get

and in this case

Aw
-=-I.
Az
Since limits are not unique, f (z) = ?! is not differentiable at z = 0 .

In fact it is nowhere differentiable. You can check it easily by proceeding as above by

considering an arbitrary point z, and observing that
7

Complex Analysis

Aw Az
and -=-
A2 Az'

Example 7: Show that the function f(z) = zn ,where n is a positive integer is

differentiable at every point in the finite complex plane.
Solution: Let z be any point in the finite complex plane.

Then,
f ( z + Az) - f(z)
. - ( z + Az)" - zn
-
Az Az

Taking limits on both sides as Az +0, we obtain

f ( z + Az) - f(z)
lim =nzn-'.
AZ+O Az
d
Therefore, - (zn)= nzn-' , where z is any point in the finite plane.
dz

Remark 1: Since f(z) = zn ,n positive integer, is differentiable at every point in the

complex plane, any polynomial Pn(z) in z is also differentiable at every point in the
complex plane.

Remark 2: Suppose that f (z) is a real-valued function of a complex variable. If f(z)

is differentiable then its derivative is zero otherwise it is not differentiable.
For, Az = Ax, where Ax is real, we have

f(z &) - f(z) = a purely real number (if exists) as Ax

+ +0
Ax
and for Az = iAy ,we have
f(z+Az)-f(z) - - f(z+iAy)-f(z)
iAz iAy

Since f (z) is differentiable lim f(z

+

+
1
= (4) ( f ( ~ i z - f(z) = a purely imaginary number.

- f(z) exists and from above discussion

Az-ro Az
we conclude that it has to be zero.

Remark 3: Although definition of differentiability of a function of complex variable

looks similar as in the case of real functions. But we will see that complex
differentiability is stronger than its real counterpart. In fact, we are going to see that if
a function is complex differentiable then its derivative is again complex differentiable.
This is not true in case of real differentiability.

You may now try some exercises to check your understanding of the concept.

Do the exercises 7,8 on page 60 of the book.

You have thus seen that a complex valued function may have derivative at every point, Analytic Functions
only at a fixed point or at no point. The natural curiosity is then whether there is any
formula or method to check differentiability with ease. A partial answer to your
curiosity is given by the famous Cauchy-Riernann equations (C-R equations), which
are a pair of equations involving first order partial derivatives. Let us see what these
equations are.

1.4 CAUCHY-RIEMANN EQUATIONS

Read Secs. 20,21,22 Chapter 2 of the book from pages 60-68. Go through
Examples 1,2 (Sec. 20), Examples 1,2 (Sec. 21) and Examples 1,2 (Sec. 22).

Let us once again look at the Theorem on Page 62, Sec. 20, of the book. Here we are
giving another proof of the theorem for your reference.
Let f(z) = u(x, y) + i v(x, y) and fl(z) exists at z, = x, + iy, .
Keeping yo fixed, we write

Since f '(x, lim (3'

+ iy, ) exists therefore, x+xo YO)- ('0 9 YO) and
X-Xo

lim v(X'yO)- V(XO'YO)both exists. In other words partial derivatives ux(x,, yo)
x+xo X-Xo

and vx(x, , yo) exist and we have

f1(z0)= ~ ~ ( ~~o ) 0. + i9v , ( x o , ~ o ) (2)
Repeating the above process by keeping 'x, fixed we obtain that
lim ~ ( ~ Y0 )3- U ( X O ~ Y
=Ol u)y ( x O ,y o ) and
Y-YO i ( -YO)
~ i

lim i[v(xo, Y)- ~ ( ~ ~009)=vy(xo,

l
YO)'
Y+YO i(y -yo)
1
Moreover, f1(z0)=:uy(x0, Y ~ ) + ~ ~ ( x ~ , Y ~ )
1

3 f'(z0) = vy(x,, yo)-i uy(x0, YO). (3)

Eqns-(2) and (3) 3 u,(xo, YO)= vy(x,, YO) and uy(x,, YO) = -v,(x,, yo)
which are the C-R equations at the point (x,, y o ) .

After going through Sec. 20 and Example 1on Page 63 of the book you must have
noticed that if a complex function is differentiable at a point then at that point C-R
equations must be satisfied by the real and imaginary parts of the function. However,
these equations are merely necessary conditions they do not guarantee the
differentiability of a complex function at a point as illustrated in the following
example.

Example 8: Let f(z) = u(x, y) + i v(x, y)

where u(x, y) = and v(x, y) = 0 .
Show that C-R equations are satisfied at z = 0 , but f is not differentiable at z = 0 .

Solution:
av av al al
Obviously, -= -= 0 and -= -= 0 , since u(x, y) and v(x, y) are
ax ay ax ay
constants (taking value 0) along real and imaginary axes or,
Complex Analysls
ux = lim U(Xy - U(o' O) = 0 etc. Thus C-R equations are satisfied at z = 0 .
x-bo X
On the other hand,
f(z)-f(0) =- m
z-0 x+iy
Putting x = r cos 8 and y = r sin 8 in the above equation, we find that
f(z) - f(o) - Ji.'EosesUlel=d m .
e-ie
(4)
z-0 r(cos8+isin8)
The right hand side of Eqn.(4) is independent of r .

If we consider the line x sin 8 - y cos 8 = 0 and keep 8 fixed (see Fig. 4),

then f(z) - f(") remains constant equal to , / m e - " for points arbitrarily
z-0
close to 0 . If 8 = 0 or 8 = R / 2 the constant value is 0 but for 8 = R / 4 this constant
1-i
takes the value -.
/ O X
2
Therefore, lim f(Z)- f(o) does not exist. Hence f(z) is not differentiable at z = 0 .
Fig. 4 z-bo z-0
-x-

Thus the C-Requations are not sufficient for the existence of derivative of a
complex function. At this juncture it is clear that we need some additional
hypothesis(es). Surprisingly what we require is nothing more than the continuity of
the partial derivatives in some neighbourhood of the point at which differentiability is
required. This fact has been stated in the theorem on Page 63, Sec. 2 1. Proof as given
on page 64 is straight forward except Eqn. (2) which needs some explanations. In
fact, derivation of Eqn. (2) needs an elementary argument of two-variable calculus
(the mean-value theorem). We hope the explanation given below will make your task
easier.
Consider
Au=u(x, +Ax, yo +Ay)-u(xo,yo)
=(u(x0 +AX, yo + Ay) - U ( X ~ , Y+OAY))+(U(XO,YO
+AY)-u(xo,Yo))

=Ax-(x,
a,l +8Ax,yo +Ay)+Ay-(x,,
all yo +$Ay), where 8 , $ ~ [ 0 , 1 ]
ax dy
[by the Mean Value Theorem of 2-variable calculus].
Since partial derivatives are continuous we may deduce that

Au = Ax-
a,l + Ay-a,l + (&,Ax+ s2Ay).
or,
ax dy
E AX E AY
We can replae E, by
Jm
then, A u = A x ( ~ ) + A Y ( $ ) + ~ h d . , ' + ( b y ) '

which is Eqn.(2) on page 64.

***
'
Sec. 22 on page 65-66 of the book introduces you to the polar form of C-R equations
in polar coordinates (r, 8) . This change of coordinate sometimes make computational
part easier. We shall illustrate it through the following example.
 Example 9: Show that f(z) = 11z4 is differentiable at z # 0 . Analytic Functions
i
Solution: Consider z = x + iy where x = r cos8, y = r sin 8 (r > 0) .
Now z = r e i e . Thus f(z) is written as f(z) =,e 448
.
r
,1
Clearly, u(r, 8) = cos 48 and v(r, 8) =,sin
r
-1
r
48 .
( u(r, 8) and v(r, 8) are defined everywhere, they are continuous and differentiable at
every non-zero point z = reie).

Now we compute partial derivatives:

C-R equations in polar
- 4cos48 - 4 sin 48
U, = , u, = form r and 0 are
r r4
v, =-
4 sin 48
, v, =
- 4 cos 48
r r4
We have
- 4cos 48 - 4 sin 48
ru, = = v e and ue =
r r4
Therefore C-R equations are satisfied at every non-zero point z = reie and all partial
derivatives with respect to r and 8 are continuous. Hence f(z) is differentiable
everywhere when z # 0 . The derivative f '(z) is given by
f ' ( ~ ) = e - (u,
' ~ +iv,)

You may now try the following exercises.

Do the exercises l(c), l(d), 2(a), 2(d), 3(c), 4(c) on page 68 and exercises 5,
6,10(a), lo@) on page 69 of the book.

We would also like you to try the following exercises.

-

E4) Show that the function f (z) = x3 + 3xy2 + i (y3 + 3x2y) is differentiable only at
points that lie on the coordinate axes.

E5) Let f ( ~ ) = ( l n r ) ~ - Q ~ + i 2 8 ( lwherer>O

nr) a n d - n c 8 S n . Showthat f is
differentiable for r > 0 , - x c 8 c x and find f '(z) .

We now introduce you to a concept that is central to complex analysis.

1.5 ANALYTICITY
It is seldom of interest to study functions that are differentiable at only a single point.
Complex functions that have a derivative at all points in a neighbourhood of a given
 Complex Analysis point deserve further study. This section is about these functions called analytic
functions.

Read Secs. 23 and 24 of Chapter 2 of the book from pages 70-73. Go

through the solved Examples 1-3 given in Sec. 24.

After going through these sections you must have noticed that being analytic at a point
is much stronger condition then having derivative at that point. A function may have
derivative at a single point without being analytic there. Look at f (z) = I z l Z
(Example 2, Sec. 18 on page 55 of the book). It is differentiable only at z = 0 but it is
not analytic at any point.
Theorem on page 71 of the book sounds familiar: If the derivative of a function is
identically zero in a domain D then the function is constant throughout D . At this
point, we would like to point out the importance of connectedness of D . This
theorem will not be true if the domain D is merely an open set. For example, if the
open set D is the union of two disjoint open discs Dl and D z as shown in Fig. 5.

Fig. 5

then the function f (z) defined by

1 if Z E D ,
f (z) =
2 if Z E D ,
has zero derivative throughout D, but f (z) is not a constant function in D .
***
To make you comfortable with the contents of Sec.23 we give below some examples.
Example 10 is infact, a simple version of Theorem on page 71 of the book.

Example 10: If f '(2) = 0 everywhere in a disk D = {z :I z I I R) ,then f (z) must be

constant throughout D .
Solution: If f (z) = u(x, y) + i v(x, y) . Then f '(2) = 0 implies
a l a v a l -a v - 0 , at every point of D .
-=-=---- [using C-R equations]
h d y d y h
Let p = a + ib and q = c + id be points in D (see Fig. 6). Then at least one of
p, = a + id and q, = c + ib lies in D because
p , q ~ ~ = 3 a+ bZ Z< R Z ,cZ+ d Z < R Z
- a Z + b Z + c Z+ d Z < 2 R Z
3 (aZ + d Z )+ (cZ+ b Z )< 2RZ
I either a Z+ d Z< R Z or, cZ+ b Z < R Z (at least)
~ i g6. either p, E D or, q, E D .
Suppose p, E D ,then f (x) = u(x, d) and g(x) = u(a, y) are real functions with zero
derivative and so they are constant. Thus u(a, d) = u(a, b) = u(c, d) and similarly
v(a, b) = v(c, d). Thus f(z) is constant [u(a, b) = u(p), u(c, d) = u(q)] .
-X-
 Example 11: The function f (z) = x + y2 + 2i xy is nowhere analytic. Analytic Functions

Solution: We identify the function u(x, y) = x 2 + y2 and v(x, y) = 2xy. The

equation u, = v, becomes 2x = 2x ,which holds everywhere. But the equation
u, = -v, becomes 2y = -2y, which holds only when y = 0 . Thus f(x) is
differentiable only at points that lie on the real axis. However, for any point
z, = x, + i 0 on the real axis and any 6 -neighbourhood of z, , the point
z, = x, + i 6 12 is a point at which f is not differentiable (see Fig. 7). Therefore, f
is not differentiable in any neighbourhood of z, and consequently, it is not analytic at
I zo.
-X-

Example 12: Suppose that f (z) is analytic in a domain D . Prove that f(z) is I
constant if Re f is constant. Fig. 7
Solution: Let f(z) = u + iv . Then Ref = u . It is given that u is constant therefore
u, = u, = 0 at every point of D . Since f is analytic in D therefore it satisfy C-R
equations. Therefore,
U, = v y and u, =-v,
a v, =Oand v, = O ateverypointof D
Now f l ( z ) = u , + i v, =-i u, + v , = O V Z E D .
Then by Theorem on page 71 of the book, f is constant.
-X-

Before we ask you to try some,. exercises on your own we make the following remark.

Remark: Failure of C-R equations signals non-analyticity. You have already seen
that a non-constant real-valued function defined in a domain is non-analytic. As a
consequence, many functions derived from an analytic function become non-analytic.
For example, none of I f 1, Re f , Im f is analytic anywhere.

You can work on the following exercises from the book.

Do the exercises 2@), 3,4(c), 6,7@) on pages 73-74 of the book.

We now move on to harmonic functions in the next section.

1.6 HARMONIC FUNCTIONS

This section deals with real-valued functions of two real variables which satisfy the
famous Laplace's equation. This equation is of fundamental importance in the
mathematical modelling of 2-dimensional physical problems concerning fluid flow,
steady heat conduction, electrostatics, and other phenomenon.

Read Sec. 25, Chapter 2 of the book from pages 75-78. Read examples 1-5
carefully. Go through Theorems 1-2 carefully.

Because harmonic functions satisfy Laplace's equations in two dimensions, they occur
widely in applied mathematics. Here we shall discuss in brief a few physical
situations where these functions occur. In Unit 6 we shall discuss in detail some more
applications of harmonic functions.
Complex Analyels Harmonic functions are solutions to many physical problems. Applications include
2-dimensional models of heat flow, electrostatics, and fluid flow etc. First we
demonstrate how harmonic functions are used to study fluid flows. We assume that an
incompressible, frictionless fluid flows over the complex plane and that all cross
sections in planes parallel to the complex plane are the same. Situation such as this
occur when fluid is flowing in a deep channel. Suppose we have a solution
4 = $(x, y) of Laplace's equation in two-dimensions and suppose that 4 is
expressible as Re w for some analytic function w . Therefore 4 possesses a
harmonic conjugate ty = Im w ;which also satisfies the Laplace's equation.

In the context of 2-dimensional fluid flow, 4 is the velocity potential and tp is the
stream function. The function w = 4 + ity is called the complex potential. Its
derivative
V(z) = w'(z) = 4, (x, y) + itp, (x, y) = -
ax ax
determines a vector field which models the velocity of fluid motion (see Fig. 8).

Obstacle

,
Fig. 8: The vector field V(x, y) = Q, (x, y) + iy (x, y) ,which can be considered as a fluid flow

It can be seen, using elementary vector calculus, that each curve 4 = constant is
orthogonal to each curve ty =constant. The latter curves follow the direction of the
fluid flow and are known as stream lines. 4 = constant are called Equipatentials.
(see Fig. 9).

I
I I ,
Equipotential

\ \ '
Streamlines
Fig. 9: The families of orthogonal curves ~ ( xy)
, = const and y(x, y) = const
 Let us consider the following example.

Example 13: Show that the harmonic function $(x, y) = x 2 - yZ is the scalar
potential function for the fluid flow

I
V(x, y) = 2x - i2y.
Solution: The fluid flow can be written as
-5

V(X, y) = wr(z) = zx + i ~ =yZz

Now w(z) = z2 and the real.part of w i.e. Re w is the desired harmonic function
$(x, y) = x2 - y 2 = Re {(x2 - y 2 )+ i2xy ). Now observe that the hyperbolas
$(x, y) = x2 - y2 = c are the equipotentials and the hyperbolas y(x, y) = 2xy = c,
are the streamline curves. These curve are orthogonal, as shown in Fig. 10.

Fig. 10: The fluid flow V(x, y) = 2x - 21 y

( I:
Let us consider the function w(z) = u z + - (1 z ( > a) where u is a positive real

number. You can easily verify that this function is analytic. Thus, its real and
imaginary parts are harmonic. These are given in polar form by

[
cose and y(r, 0 ) = u r-- ar2)sin0.

We have y = 0 for 0 = 0 and 0 = sc ,as well as on the circle r = a . For every large r ,
the streamlines approximate to straight lines parallel to x -axis (real). This example
models uniform 2-dimension fluid flow parallel to the x -axis, into which a circular
cylinder of radius 'a' has been placed (see Fig. 11).

Fig. 11: Flow past a cylinder

Complex Analysis The complex potential has many physical interpretations. Suppose that we have
solved a problem related to steady state temperature (see Example 1, page 75,
Sec. 25); then a similar problem with the same boundary conditions in electrostatics is
obtained by interpreting the isothermals as equipotential curves and the heat flow lines
as fluxlines. In fact, this implies that heat flow and electrostatics corresponds directly.
Below we give a chart which gives various interpretations of the families of level
curves Q(x, y) = const and ~ ( x y) , = constant.

Physical Phenomenon 4 (x, y) = constant y(x, y) = constant

Heat flow Isothermals . Heat flow lines
Electrostatic Equipotential lines Fluxlines
Fluid flow Equipotentials Streamlines
Elasticity Strain function Stresslines
Diffusion Concentration Lines of flow
Current flow Potential Lines of flow
Magentism Potential Lines of flow

Two-dimensional Electrostatics

A 2dimensional electrostatic field is produced by a system of charged wires, plates

and cylindrical conductors that are perpendicular to the z -plane. Now this induces an
electric field E(x, y) that can be visualized as force acting on a unit positive charge
placed at the point (x, y) . In the study of electrostatics the vector field E(x, y) is
shown to be conservative and derivable from the function Q(x, y) ,called the
electrostatic potential, as expressed by the equation:

The level curves $(x, y) = c, are called the equipotential curves, and the curves
y(x, y) = c, are called the lines of flux. For example, if a small charge is allowed to
move under the influence of E(x, y) ,then it will travel along a line of flux. To be
more specific let us consider the following situation.
We find the electrical potential Q(x, y) in the region between two infinite coaxial
cylinder r = a and r = b ,which are kept at the potential u, and u 2 , respectively (see
~i~.12).'
The function w = log z =.In I z I +i arg z maps the annular region between the circles Analytic Functions
r = a and r = b onto the infinite strip In a < u < ln b in the w -plane as shown in
Fig. 12. Now potential @(u,v) in the infinite strip will have the boundary values

$(lna, v ) = u 2 and @(lnb,v)=u, for all v .

We are seeking a solution 4 (u, v) that takes on constant values along the vertical
lines u = In a and u = In b with 4 (u, v) being a function of u alone, that is,
I$ (u, v) = p(u) for, In a I u I In b and for all v

Now Laplace equation, 4, (u, v) + 4, (u, v) = 0 .

3 pa(u) = 0 p(u) = mu + c where m and c are constants.

,
The boundary conditions 4 (In a, v) = p (In a) = u, and 4 (In b, v) = p(ln b) = u lead to
the solution
4(u, V)= U I + U2-U1 (u-lna)
lnb-lna
Since u = In 1 z I ,we conclude that the potential @(x,y) is

(I)(%Y)=u1 + U2 (lnlzl-lna).
lnb-lna

In Fig. 12 you can see that equipotential curves $(x, y) = const are concentric circles
centred at the origin and the line of flux (dotted lines) are portion of rays emanating
from the origin.

Do the exercises l(c), 2,5,7,9 on pages 78-80 of the book.

1.7 SUMMARY
In this unit we have covered the following:

1) Complex numbers can be defined as ordered pairs (x, y)' of real

numbers and represented as points in the complex plane with rectangular
coordinates x and y .

2) Addition and multiplication properties of complex numbers are the same as for
real numbers.

3) I I
The real positive number ( z = x + i d = Jx2 + y 2 is called the modulus or the
absolute value or the magnitude of the complex number z = x + i y .

4) The distance between two complex numbers 2, = (x, + iy, ) and

z 2 = ( x 2 + i y 2 ) isgivenby

d = ( z 2 -2, J=l(x2-x,) + i ( y 2 - y , ) l = d ( x 2 - x , ) ~+(y2 +y,12 .

5) For any two complex numbers z,, z, ,we cannot write z, < z2 or, z, > z, . We
can however compare the magnitudes of two complex numbers.
Complex Analysis 6) Complex roots of a polynomial equations P(x) = 0 with real coefficients occur in
conjugate pairs.

7) For two non-empty sets S, and S, of complex numbers, a rule f which assigns a
complex number w is S, for each z in S, , is a complex valued function of a
complex variable z and is written as w = f(z).

8) The value of w for a given z = z, ,that is f (z,) is called the image of the point
z = z, under the rule f .

9) Mapping or transformation is a way of displaying information about the function

w = f (z), z = (x, y) and w(u, v) ,by drawing the z and w planes separately.

,
10) Limit of f (z) as z approaches z, is a number w ,or that lim f (z) = w ,
z+zo
,
,
means that the point w = f (z) can be made arbitrarily.closeto w if we choose
the point z close enough to zo but distinct fiom it.

1 1) A function f is continuous at a point z, if lim f(z) = f(zo) .

z-ZO

12) The function f is differentiable at z, if lim f(Z) - f(zO) exists. This limit is the
z+zo Z-Zo

derivative of f(z) at the point z = z, and is denoted by f '(2,).

13) Cauchy-Riemann equations are the pair of equations

u x ( x ~ , ~ ~ ) = v ~and ( xu, ~(x,,y,)=-v,
, ~ ~ ) (x,,y,), whichthe first-order
partial derivatives of the component functions u and v of a function
f(z) = u(x, y) + i v(x, y) must satisfy at a point z, = (x,, yo) when the
derivative of f exists there.

14) Cauchy-Riemann equations provide necessary conditions for the existence of

f '(z,) but their satisfaction at a point z, is not sufficient to ensure the existence
of the derivatives of a function f(z) at that point.

15) A function f(z) is analytic at a point z = z, if it is defined, and has a derivative,

at every point in some neighbourhood of z, . It is analytic in a domain D if it is
analytic at every point in D .

16) If f(z) = u(x, y) + iv(x, y) is analytic in a domain D then u(x, y) and v(x, y)
satisfy Cauchy-Riemann equation at each point in D .Conversely, if the first
order partial derivatives of u(x, y) and v(x, y) are continuous and satisfy the
Cauchy-Riemann equations in a domain D ,then f(z) = u(x, y) + iv(x, y) is
analytic in D .

17) A real vauled function H(x, y) of real variables x and y is harmonic in a

domain D of the xy -plane if, throughout that domain, it has continuous partial
derivatives of the first and second order and satisfy the Laplace equation
H, (x, y)+H,(x, y)=O.

18) The real and imaginary parts of an analytic function are harmonic functions.
 Exercises on page 35 of the book.
Analytic Functions
I:
Z
1
I
1. (c) f (2)'= -
z + z . Clearly domain of definition contains those complex numbers
z forwhich z + Z # O = 2 R e z#O*Re ~ $ 0 . Y4
Thus domain of definition is (z :Re z # 0).

Geometrically speaking it is whole complex plane without imaginary axis (see

Fig.13). This set is open but not connected. You can not join two points say Real axis
z,, z2,one lying to the left and other to the right of imaginary axis, by a
polygonal line without crossing the imaginary axis.

1
I
Imaginary
axis $
(d) f (2) = -. Again it can be seen easily that the domain of definition
1-1zI2 Fig. 13
consists of those complex numbers for which I z (#1 (i.e. 1- ( z l2 + 0 ). Thus
domain of definition is (z :1 z I# 1) . Geometrically, it is whole complex plane
minus the unit circle centred at the origin (see Fig. 14). It is also open but not
connected for obvious reasons.

In polar coordinates: (r # 0)

2 f (reie)= r (cos8 + i sin 8) + -1 (cos(-8) + i sin(+)

r

Fig. 14

Exercises on pages 42-43 of the book.

4. w = exp z
j ~ + i v = e ~=ex(cosy+isiny)
+'~
j u = e X cosy, v = e Xsiny
j u 2 + v2 = e Z Xand t a n y = v / u 3 y = t a n - ' ( v / u )
a n d x = l n ( u 2 + v2 )112 .
Now ay = x(a # 0) is transformed to a tan-' (v / u) = ln (u2 + v 2 )112
(,2 + v 2 ) ~ ~-2e a m - l ( ~which
~ ~ ) reduces to p = ea4 (equation of
spiral) in polar form where, p = (u2 + v2)'I2, $ = tan-' (v / u) .

5. Go through Example 1, Sec.13 once again.

Horizontal segment AB will be mapped onto the portion of ray $ = c lying

between p = ea(A') and p = eb(B') . The images of horizontal segments above
AB and joining the vertical parts of the boundary are portions of ray moving left
 Complex Analysis of A' B' ; eventually, the images of the line segment DC is the portion of ray
4 =d lying between p = ea(Dl) and p = eb(c') (see Fig. IS). Thus image of the
rectangular region ABCD (a Ix I b, c I x <_ d) is the region
ea s p < e b , c c $ < d (sameasFigure21, page41).

vA

w=eL C'
A
'--' ' .

B'

X
* 0
b
u

Fig. 15

Exercises on pages 53-54 of the book.

4. We observe that for n = 1

limz=zo.
z+zg

Therefore it holds for n = 1. Suppose it holds for n = k i.e.

limz k = z ko .
z+zo

Now, lim zk+'= lim (zk.z)

Z-+zo 2420

~
I
=
)
zk
Z) (using Property (9) Sec.1S)

= 2:. zo (using Eqns. (5) and (6))

= z kot 1 .

Thus it holds for n = k + 1. Hence by Mathematical induchon lim zn = z: holds

Z-PZ,,

for all positive integer n .

5. Go back to Example 2, Sec.14 for one more reading. We know that if li

z 2-0
{q
exists, it could be found by letting the point z = (x, y) approach the origin in any
manner (along any path).

When z = (x, 0) , (x z 0) then

and when z = (x, x) , (x # 0) then

( ( x+ )( )
X - ix
IX 1-1
2
=-I

Thus, by letting z approach the origin along real axis, we would find that the
desired limit is 1. An approach along line y = x ,on the other hand, gives the
 (zT
Analytic Functions
limit as - 1. Since a limit has to be unique by definition, therefore limit lirn
z+o
=z
does not exist. If we consider, letting z approaches origin along imaginary axis

then I;( 2

= (%)2

= 1 , it is the same limit as we get along the real axis. This

was the reason for considering another path namely y = x .

9. We are given that lirn f(z) = 0 . By the definition of limit, for a given E > 0 there
z+zg

exists 6 , >O suchthat )f(z))=If(z)-O)<E,whenever O<Iz-z, ( < 6 , .

If 1 g(z) I IM (M a positive number) for all z in some neighborhood of z, , it
means there is 6, > 0 such that ( g(z) IIM Vz such that I z - z, (<6, .
i Now choosing 6 = min{6,, 6,), we have

If(z)g(z)-O(=If(z)g(z))=If(z)I (g(z)(< ~ M = & , , w h e n e v e O<(z-z,

r (<a.
(given E > 0, 3 E, > 0 such that E M= E, and conversely consequently).

Hence ( f (z) g(z) 1 < E, whenever 0 <1 z - z, 1 < 6

3 lirn f(z) g(z) = 0 .
z+zg

if lim -A =0
~ + oT(l/ z)
1
if lim- =0
Z-PO a + bz

c + dz
c + dz
if lim- =0
z+o a + bz
if c=O.

b) lirn T(z) = a / c and lirn T(z) = co

Z-+m z+-d / c
1
if limT(l/z)=a/c and lirn -=O
z+O z+-dlc T(z)

a+bz cz+d
if lim-=a/c and lim -- - 0
0 ' 2 c + dz Z-P-d/c az + b
if c#O

13. Suppose S is unbounded. We have to show that every neighborhood of the point
at infinity contains at least one point of S .

Let there be a neighborhood of the point at infinity containing no point of S . That

is, 3 a small E > 0 such that 1 z 1 > 11E contains no point of S . It means
S {z :( z I_< 11E) 3 S is contained in ( z I < 1/ E i.e. S is bounded by the circle
1.2(=1/ E 3 S is bounded. It is a contradiction. Therefore every neighborhood of
the point at infinity contains at least one point of S .

Conversely, suppose every neighborhood of the point at infinity contains at least

one point of S . Let S be bounded. Then by definition of boundedness 3 a
positive number R such that S lies inside the circle I z 1 = R . Now choosing
Complex Analysis E = 1/ R ,we see that I z ( > 1/ E contains no point of S . It is again a contradiction.
Therefore S is unbounded.

Exercises given in the unit on page 10.

El) a) f(z)=1zl2=x2+ y 2 , U ( X , ~ ) =+Xy 2~, v(x,y)=O

u(x, y) and v(x, y) are continuous for all x, y E R*. Therefore f (z) is
continuous for all z in the complex plane.

b) Do it yourself.

b) On imaginary axis.

Exercises on page 60 of the book.

7. Wewritem=-n. T h e n f ( z ) = z n = z - m = L ( w h e r e m > ~ a n d z +Now

~).
zm
using the formula for the derivative of a quotient of two functions and
d
-zn (where n is positive integer), we have
dz

- -
Aw (z+Az)-E S+Az-E Az
8. a) Here -=
Az Az
- Az =-,Az
When Az approaches the origin horizontally through the point (Ax, 0) on the
real axis:

Aw Az Aw
In this case, -= -= 1.. Hence if the limit of - exists, its value must
Az Az Az
be 1. However, when Az approaches the- origin vertically through the point
(0, Ay) on the imaginary axis, we find Az = Ax + iAy = 0 + iAy = -iAy = -Az
AW -Az
and-=- dw
= -1. Thus limit is not unique and so - does not exist.
Az Az dz

b) Do it yourself.

Aw Im(z+Az)-Imz - Imz+ImAz-Imz - Im (Az)

c) Here -=
Az Az Az Az
If Az approaches the origin horizontally through (Ax, 0) ,we get
Aw
.
Im Az = 0 In this case -= 0 . Hence if the limit exists, its value must be
Az
 0 . However, when Az approaches the origin vertically through (0, Ay) then
Aw Ay dw
Im Az = Ay and -= -= 1 . Thus -= f '(z) does not exist.
Az Ay dz

I Exercises on page 68 of the book.

i 1. c) f(z)=2x+ixy2.
Here u(x, y) =2x, v(x, y) = xy2.
Thenu,=2,uy=0,v,=y 2,vy=2xy.
If f (z) is differentiable at any point (x, y) then by Theorem page 62 of
the book.
U, = v y and u, =-v,

3 2 = 2xy and 0 = -y2

3 xy=l andy=O
which is not possible for any (x, y) . Therefore f '(z) does not exist at any
point.

d) f(z)=ex e-I, = e x (cosy-isiny).

Wehave, u = e x cosy and v=-ex siny.
Now U, = e x COSY, u, =-ex siny, v, =-ex siny and v, =-ex cosy.
i If f (z) is analytic at a point (x, y) then by C-R equations
u, = v y , u y =-vx
3 excosy=-excosy and - e x s i n y = e x s i n y
3 ex cosy=O and e x siny=O V (x, y)
3 cosy=Oandsiny=O V(x,y)
which is not possible for any (x, y) . Therefore f '(z) does not exist at any
point.
i
b
2. a) f ( z ) = i z + 2 . [z=x+iy]
This function is defined everywhere. We express f (z) as
f(z)=(2-y)+ix.
Here, u(x, y) = 2 - y and v(x,.y) = x . We observe that u(x, y) and
v(x, y) are polynomials in x and y therefore partial derivatives exists
everywhere, they are u, = 0, u, = -1, v, = 1, v, = 0 . Clearly, being constant
functions they are continuous everywhere.

Moreover, C-R equations are satisfied everywhere i.e.

U, = O = v y and u, =-1=-v,.

Thus every ca.ndition of the Theorem on page 63 is satisfied. Therefore f '(z)

exists at every point and is given by
f f ( z ) = u X+iv, = O + i . l = i .
We have shown that f '(z) exists everywhere and it is given by f '(z) = i .
Thus derivative is a constant function. Put f '(z) = g(z) .
Now g(z) = i = u(x, y) + i v(x, y)
3 u(x,y)=O, v(x, y ) = l .
Clearly being constant functions u ,,u, ,v, ,v, exist everywhere and
U, =O,uy =O,V, =O,vy = O and u, = v , and u, =-v,. Thusallthe
Complex Analysis conditions of Theorem on page 63 are satisfied. Therefore g(z) is
differentiable i.e. gf(z) exists everywhere i.e. f "(2) = gf(z) exists everywhere
and is given by f "(z) = gf(z)= u, + i v, = 0 .

More generally, if f (z) = az + b , a linear function, then f '(z) = a and

f "(2) = 0 .

d) f (z) = cos x cosh y - i sin x sinh y .

We have u(x, y) = cos x cosh y, v(x, y) = - sin x sinh y .
It is defined everywhere. We observe that u(x, y) and v(x, y) are simply a
product of trigonometric and hyperbolic functions, therefore differentiable at
every point. Thus partial derivatives exist everywhere i.e.
U, =-sinxcoshy, u, =cosxsinhy, v, =-cosxsinhy,

v, =-sinxcoshy.
For every (x, y) , u, = v, and u, = -v, . Thus every hypothesis of Theorem
on page 63 is satisfied. Hence f (z) is differentiable everywhere and f '(2)
exists everywhere. Therefore f '(z) = u ,+ i v ,.
fl(z)=-sinxcoshy-icosxsinhy.
By the similar arguments we see that ff(z) is also differentiable everywhere
with derivative
f "(2) = - cos x cosh y + i sin x sinh y = -f (z) . 1
3. c) f(z)=zIrnz.
We express f (z) as f (z) = (x + i y) y = xy + i y2 .
Thus, u(x, y) = xy and v(x, y) = y 2 .
First we try to find out those points at which C-R equations are satisfied
i.e., u, = v , and u, =-v,.
Weget y = 2 y and x = O = y=O and x=O.
Thus C-R equations are satisfied only at (0,O) ,i.e. at z = 0 . We see that
being combinations of simple functions (elementary) u and v are
differentiable everywhere, i.e., there is a neighbourhood of z = 0 where partial
derivatives exist and are continuous. Thus according to the Theorems of
Secs.20 and 2 1, f (z) is differentiable only at z = 0 with
f'(O)=uX(0)+iv,(0)=O+iO=O.

4. c) f (z) = e-, cos(1nr) + i e-, sin(1nr) (r > 0, 0 < 8 < 2 x )

We have u(r, 8) = e-, cos(1n r) and v(r, 8) = e-e sin(1n r) . It is obvious
that in the given domain u(r, 8) and v(r, 8) are defined and their partial
derivative exists and we get u, = -e -0 1 , u, = -e.-' cos(1nr) and
sin(1n r) -
r

1
and r v, = r.e-, cos(1nr)- = e-e cos(1nr).
r
Clearly these partial derivatives are continuous in the given domain and C-R
equations are satisfied. Therefore, Theorem on page 66, Sec.22 says that
 function is differentiable in the indicated domain of definition and its Analytic Functions
derivative is given by f '(2) = e-' (u, + i v, )
- e-e sin(1n r) -1 + i e-e cos(1n r)
r
i i f (z)
= -(e-e cos(1nr) + i e-' sin(1n r))= -.
r ele z

Exercises on page 69 of the book.

5. f(z) = x3 +i(l- y)' .

Here u(x, y) = x3, v(x, y) = (1 - y)' .

Now C-Requations are satisfied only if 3x2 = -3(1- y)2 and 0 = 0 which
imply x = 0 and y = 1. Thus, C-Requations are satisfied only at (0,l) ,i.e.,
z = i . Also observe that in every neighbourhood of z = i ,partial derivatives
exist and are continuous. Therefore f(z) is differentiable only at z = i . It is
therefore legitimate to write f '(2) = u, + i v, = 3x2 + i. 0 = 3x .

6. f ( ~ ) = / $ , when z # 0
(0 , when z = 0
-2 -3
Z Z
Wecanwrite -=-.
z zz
x3 -3xy2
Hence, u(x, y) = 3 (x, Y)* (OYO)

h3 / h 2
Now u ,(0,O) = lim u(h, 0) - 4 0 , 0) = lim-=l
h+O h h+O h
likewise, u, (0,O) = 0, v, (0,O) and v, (0,O) = 1 .
Thus C-Requations are satisfied at (0,O) or at z = 0 .

10. a) We express F(x, y) as follows:

F = F[x (z,z), y(z, z)] . ( x and y are treated as functions of z and Z )

Now differentiating F w.r.t Z and using chain rule we get,

b) For a function f (z) = u(x, y) + i v(x, y) we are given that u, = v, and

Complex Analysls

[using C-R equations]

af
Thus, -=o. [complex form of C-R equations]
az
Exercises on page 15 of the unit.

E4. f ( z ) = x 3 + 3 x y 2+ i ( y 3 + 3 x 2 y ) .
Here u(x,y)=x3 +3xy2 and ~ ( x , y ) += 3~x 2~y .
Now u X = 3 x 2+ 3 y 2 , u y=6xy,vx =6xy and v, =3y2 + 3 x 2 .
If f(z) is differentiable at (x, y) then it must satisfy C-R equations:
U, = v y and u, =-v,

3 3(x2 + Y 2 )= 3(x2 + Y2) [which is identically satisfied] and 6xy = -6xy

3 x y = O a x = O or y=O.
So, if f(z) has to be differentiable then the points of differentiability are only
those points which lie along coordinate axes. Since partial derivatives are
continuous everywhere, therefore f (z) is differentiable along coordinates axes,
i.e., f(z) is differentiable at points of the form (x,O) or (0, y) only.

E5. Wehave u(r,8)=(1nr)~-€I2, v(r,8)=28(Inr).

Clearly, u and v are defined in the indicated domain. We compute partial
derivatives
2 2
u,(r,0) =-(lnr), u, = -20, v, =-0, v, = 2(lnr), which exist in the domain
r r
{r, 0) : r > 0, - x < 8 < x) and are also continuous here in.
2
Also, r u , =r.-(lnr)=2(lnr)=ve
r

( )
and u, = -28 = -r - = -r v, . C-R equations are satisfied.
2re
Hence, f '(z) exist for all z in the indicated domain and

Exercises on pages 73-74 of the book.

2. b) f(z)=2xy+i (x2 - y 2 ) .
We have, u(x, y) =2xy, v(x, y) = x 2 - y 2 .
Therefore, u, = 2y, u, = 2x, v, = 2x, v, = -2y and so C-R equations are
satisfied only at (0, 0). This f (z) is nowhere analytic. Note that f '(0) exists.

d
3. Let f (z) and g(z) be any two entire functions. Then by definitions -f(z) and
dz
d
-g(z) exists everywhere in the complex plane. Clearly, fog and gof are
dz
defined in the plane and (fog) (z) = f (g(z)) Vz and (gof) (z) = g(f (2)) Vz .
 Then by chain rule: Analytic Functions

d
-
dz
(gof )(z) =
d
g(f (2)) =
d
,
d
~ ( 4
du
[where f (2) = u]

=-g(u).
du
-
dz

= gl(f(z)). fl(z) vz
Ii 3gof is analytic everywhere. Complete the remaining part yourself.

z2 +1
4. c) f(z)=
(2 + 2) (z2 + 22 + 2) '
Singular points are the solution of the equation:
(z+2) (z2 + 2 2 + 2 ) = 0
*z=-2, z = - l f i .
Since f(z) = -('" where P(z) and Q(z) are polynomials we know fmm
i Q(z)
quotient rule of differentiations that fl(z) exists everywhere except at
z = -2, - 1f i because function is not defined at these points.

6. We have g(z) = In r + i8 (r > 0, 0 < 8 < 2 ~ (Z

) # 0). Component functions are:
u(r, 8) = lnr and v(r, 8) = 8 .
We have,

Observe that partial derivatives exists in the indicated domain and are
continuous there. C-R equations are also satisfied as
r U, = l = v e and u, =O=-r v,.
Hence, g(z) is analytic in the indicated domain.
Again,
gr(z)= e-ie(u, + iv,) = e-"

1
Applying chain rule and gl(z) = - ,we get
Z

7. b) Suppose lf(z)l=c V z c D .
When c=O, then ( f ( z ) J = O a f ( z ) = OV z c D ~ f ( z isconstant.
)
When c # 0 then I f(z)I2=c2
*u2 + v 2 = c 2 [f(z) = u + iv]
Therefore,

Hence, by the C-R equations.

Complex Analysis Eliminationof u, gives (u2 + v 2 ) u x= O = c 2 u,
3u D.
,= 0 , everywhere in
Similarly, u, , v, , v, are zero everywhere in D . We deduce that f is a
constant.

Exercises on Page 78 of the book.

1. c) Given function is u(x, y) = sinh x sin y .

We observe that
a2u a2u
-+-=sinhxsiny=sinhxsiny=O V x and y ,
ax2 ay2
i.e. u(x, y) is harmonic throughout the entire complex plane. If v(x, y) is
conjugate to u(x, y) then by C-R equations:
U, = V, and u, = -v,

&
Now v, = -= cosh x sin y ,therefore
ax
v(x, y) = -cosh x cos y + +(x) and its differentiation partially w.r.t. x gives
v, = -sinh x cos y + +'(x)
3 -u, =-sinhxcosy++'(x)

3 - sinh x cos y = - sinh x cos y + +'(x)

3 +'(x) = 0
3 +(x) = C where C is an arbitrary real number
and we get,
v(x, y) =-coshxcosy+ C .
For C = 0 , we get a particular harmonic conjugate v(x, y) = - cosh x cosy .

2. Since v and V if exist are harmonic conjugate of u in a domain D ,

f(z) = u + iv and f, (z) = u + iV are analytic functions in D . Now by C-R
equations:
U, = v y , U, =-V, and u, =V,, u, =-V,
3 V, =Vy and v, =V,
3 (v-V), = O and (v-V), = O
3 v-V=const.
i.e. v = V + C where C is a constant.

5. We have C-R equations in Polar-form as:

r u,=v, (0
u, = -r v, . (ii)
Differentiating (i) w.r.t r ,we get
U, + r U, = v,, . (iii)
Differentiating (ii) w.r.t 8,we get
u,, = -r v,. (iv)
Multiplying (iii) by r and adding to (iv) we get
r U , + r 2 U,+Uee=O.

7. As given in suggestion:
slope of tangent at z, = (x, , yo) to curve u(x, y) = c, is given as
m &fax
- &fay
 and slope of tangent to curve v(x, y) = c, is given by

- -
NOWmlm2 = a x ' & =-1 (u, = v y and u, =-v,)
alav

3 u(x, y) = c, and v(x, y) = c, are orthogonal at z, .

1
9. We are given f (z) = - . It is an analytic function throughout the plane except at
z
point (0, 0) i.e. at origin.

We can express f (z) = 1/ z as:

x-iy x
f (2) = 3 f (2) = -i Y
xZ + y2 xZ + yZ x Z + y Z'
Here u(x, y) =
X
and v(x, y) = - Y
x Z + yZ x Z+ y 2 '

Now families of level curves u(x, y) = c, and v(x, y) = c, where c, and c2 are
arbitrary real constants. (f(z) # 0 for any z E D).

Now,
X

x 2 + yZ
= c I and -Y ,
x Z+ y
= c 2 or Y = -C2.
xZ + yZ

and if c, # 0, =-c2ax
xZ + yZ

Families of level curves are shown in Fig.16.

Fig. 16

Remember v(x, y) = c2 and u(x, y) = c, are circles centred at y -axis and

x-axis, respectively and passing through the points (0, 0) but this point is
 Complex Analyds removed in all cases. For c, = 0 we get y-axis (minus origin) and for c, = 0 ,we
get x-axis (minus origin).

Check the hnctions u(x, y) = c, and v(x, y) = c , for orthogonality yourself.

-0-

34